the autism-spectrum quotient (aq)—adolescent version
TRANSCRIPT
The Autism-Spectrum Quotient (AQ)—Adolescent Version
Simon Baron-Cohen,1,2 Rosa A. Hoekstra,1 Rebecca Knickmeyer,1 and Sally Wheelwright1
The Autism Spectrum Quotient (AQ) quantifies autistic traits in adults. This paper adaptedthe AQ for children (age 9.8–15.4 years). Three groups of participants were assessed: Group 1:n=52 adolescents with Asperger Syndrome (AS) or high-functioning autism (HFA); Group 2:n=79 adolescents with classic autism; and Group 3, n=50 controls. The adolescents with AS/
HFA did not differ significantly from the adolescents with autism but both clinical groupsscored higher than controls. Approximately 90% of the adolescents with AS/HFA and autismscored 30+, vs. none of the controls. Among the controls, boys scored higher than girls. The
AQ can rapidly quantify where an adolescent is situated on the continuum from autism tonormality.
KEY WORDS: AQ; adolescents; screening; autistic spectrum; Asperger Syndrome.
INTRODUCTION
In an earlier issue in this journal, we reportedon the Autism Spectrum Quotient (AQ) in adultswith high functioning autism (HFA) or AspergerSyndrome (AS) (Baron-Cohen, Wheelwright, Skin-ner, Martin, & Clubley, 2001). The adult AQ wasdeveloped because of a lack of a quick and quan-titative self-report instrument for assessing howmany autistic traits any adult has. The minimumscore on the AQ is 0 and the maximum 50. If anadult has equal to or more than 32 out of 50 suchtraits, this is highly predictive of AS. The AQ hasbeen found to correlate inversely with the EmpathyQuotient (EQ) (Baron-Cohen & Wheelwright, 2004),the Friendship and Relationship Quotient (FQ)(Baron-Cohen & Wheelwright, 2003) and to corre-late positively with the Systemizing Quotient (SQ)
(Baron-Cohen, Richler, Bisarya, Gurunathan, &Wheelwright, 2003).
The AQ has also been found to be stronglypredictive of who receives a diagnosis of AS in a clinicsetting (Woodbury-Smith, Robinson, & Baron-Cohen,2005). The AQ also reveals sex differences (males>females) and cognitive differences (scientists>non-scientists) (Baron-Cohen et al., 2001), a pattern ofresults that has been closely replicated in a Japanesesample (Wakabayashi, Baron-Cohen, & Wheelwright,2004). This latter pattern suggests that these effects arenot culture-specific and may instead reflect sexualdimorphism in the brain and differences in neuralorganization between scientists (a clear example of‘systemizers�) and non-scientists.
The AQ depends on self-report, which may be aconcern in individuals whose social deficits mayimpair their accuracy in self-awareness. However, aparent-version confirms that adults of normal IQwith autism spectrum conditions are able to providesuch information reliably (Baron-Cohen et al., 2001).Although the AQ is useful both clinically and inresearch studies as a screen for diagnosis, it has notbeen studied as a general population screen, andindeed if it was tested for this purpose it is likely that
1 Department of Psychiatry, Autism Research Centre, University
of Cambridge, Cambridge, UK.2 Correspondence should be addressed to: Simon Baron-Cohen,
Department of Psychiatry, Autism Research Centre, University
of Cambridge, Douglas House, 18b Trumpington Road, CB2
2AH, Cambridge, UK.; e-mail: [email protected]
Journal of Autism and Developmental Disorders, Vol. 36, No. 3, April 2006 (� 2006)
DOI 10.1007/s10803-006-0073-6
Published Online: March 22, 2006
3430162-3257/06/0400-0343/0 � 2006 Springer ScienceþBusiness Media, Inc.
a significantly higher cut-off would need to beemployed to keep false-positives to a minimum.
Given the usefulness of the AQ, it is of interest totest a revised version of this instrument with adoles-cents. In this paper, we test (1) whether similar resultsare found on an adolescent version of the AQ, and (2)whether similar results are found in classic autism as inAS. Whereas the adult AQ (for ages 16+) entails self-report, the adolescent AQ requires a parent/carer tocomplete it, but otherwise retains the same items andstructure as the adult version. The adolescent AQ wasdesigned to be short, easy to use, and easy to score.
The adolescent version is shown in Appendix 1.It comprises 50 questions, made up of 10 questionsassessing 5 different areas: social skill (items1,11,13,15,22,36,44,45,47,48); attention switching(items 2,4,10,16,25,32,34,37,43,46); attention to detail(items 5,6,9,12,19,23,28,29,30,49); communication(items 7,17,18,26,27,31,33,35,38,39); and imagination(items 3,8,14,20,21,24,40,41,42,50). Each of the itemslisted above scores 1 point if the respondent recordsthe abnormal or autistic-like behaviour eithermildly or strongly (see below for scoring each item;Abnormality—poor social skill, poor communicationskill, poor imagination, exceptional attention todetail, poor attention-switching/strong focus of atten-tion). Approximately half the items were worded toproduce a ‘disagree� response, and half an ‘agree�response, in a high scoring person with AS/HFA.This was to avoid a response bias either way.Following this, items were randomized with respectto both the expected response from a high-scorer, andwith respect to their domain.
Participants
3 groups of participants were tested:Group 1 comprised n=52 adolescents with AS/
HFA (38 males, 14 females). This sex ratio of 2.7:1(m:f) is similar to that found in other samples (Klin,Volkmar, Sparrow, Cicchetti, & Rourke, 1995). Allparticipants in this group had been diagnosed bypsychiatrists using established criteria for autism orAS (APA, 1994). They were recruited via severalsources, including the National Autistic Society (UK),specialist clinics carrying out diagnostic assessments,and adverts in newsletters/web-pages for children withAS/HFA. Their mean age was 13.6 years (SD=2.0,range 10.3–15.4). They all attended mainstreamschooling and by parental report, had an IQ in thenormal range. See below for a check of this. Becausewedid not collect data on age of onset of language these
individuals are grouped together, rather than attempt-ing to separate them intoAS vs.HFA. The final sampleof 52 comprises those who responded from a largersample of 63. All participants were regarded asindependent, in being genetically unrelated.
Group 2 comprised n=79 adolescents with clas-sic autism (63 males, 16 females). Again, this sex ratioof 3.9:1 is similar to that found in other samples. Allparticipants in this group had been diagnosed bypsychiatrists using established criteria for autism(APA, 1994). They too were recruited via theNational Autistic Society (UK), specialist clinicscarrying out diagnostic assessments, and adverts innewsletters/web-pages for children with autism. Theirmean age was 12.5 years (SD=1.7, range 9.8–16.0).They all attended special schools for autism orlearning difficulties, and by parental report, had anIQ below the normal range. See below for a check ofthis. The final sample of 79 comprised those whoresponded from a larger sample of 85.
Group 3 comprised 50 adolescents selected atrandom (n=25 males and 25 females). They weredrawn from 200 adolescents. They were all attendingmainstream schools (2 primary and 2 secondary) inthe East Anglia area. Questionnaires were distributedby the schoolteachers via the children�s school class.Their mean age was 13.6 years (SD=1.8, range 10.1--16.5). The 3 groups did not differ significantly at thep=.05 level for age.
In Groups 1 and 2, 15 individuals were randomlyselected from the individuals who had returned anAQ and invited into the lab to check pro-rated IQ,using 4 subtests of the WAIS-R (see below). Theseparents were also asked to complete a second AQ as ameasure of test–retest reliability.
Method
Parents were sent the AQ by post, and wereinstructed to complete it as quickly as possible (toavoid thinking about responses too long). To confirmthe diagnosis of Group 1 being high-functioning andthe diagnosis of Group 2 being lower-functioning, 15of each were randomly selected and invited into thelab for intellectual assessment using 4 subtests ofthe WISC-R (Wechsler, 1958). The 4 subtests of theWISC-R were Vocabulary, Similarities, BlockDesign, and Picture Completion. On this basis, all15 participants selected from Group 1 had a proratedIQ of at least 85, that is, in the normal range(mean=106.5, SD=8.0). These participants� parentswere also given the AQ when they came into the lab
344 Baron-Cohen, Hoekstra, Knickmeyer, and Wheelwright
in person, in order to investigate test–retest reliabilityfor the instrument on a small subset of children. All15 of the participants selected from Group 2 had aprorated IQ of below 84, that is, below the normalrange (mean=72.6, SD=8.0). Note that we haveconservatively used an IQ of >85 as an inclusioncriterion for AS/HFA.
Scoring the AQ
‘Definitely agree� or ‘slightly agree� responsesscored 1 point, on the following items: 2, 4, 5, 6, 7, 9,12, 13, 16, 18, 19, 20, 21, 22, 23, 26, 33, 35, 39, 41, 42,43, 45, 46. ‘Definitely disagree� or ‘slightly disagree�responses scored 1 point, on the following items: 1, 3,8, 10, 11, 14, 15, 17, 24, 25, 27, 28, 29, 30, 31, 32, 34,36, 37, 38, 40, 44, 47, 48, 49, 50. If a respondent leftmore than 5 items blank, this AQ was deemed to beincomplete and was omitted from the sample.
Results
Mean AQ scores (total) for each group, brokendown by sex and by subdomain, are shown inTable I. Comparing groups using an ANOVA of
Total AQ score by GROUP and SEX, we found aspredicted a significant effect of GROUP (F (2,175)=201.49, p<.001). Post Hoc Scheffe testsrevealed that the two clinical groups scored signif-icantly higher than the control group (p<.0001), butthat the two clinical groups did not differ from eachother. The main effect of SEX was not significant (F(1, 175)=2.84, p>.09), but there was a significanttwo-way interaction of GROUP�SEX (F (2,175)=6.48, p=.002).
T-tests confirmed that there was a significant sexdifference (t=)3.27, p=.002) in the control group(males scoring higher than females), confirming thesame effect reported with the adult AQ. There wereno significant sex differences in the clinical groups(group 1: t=1.90, p=.063; group 2: t=)2.0, p=.049(non-significant when Bonferroni correction formultiple tests is used)). The clinical groups differedfrom the control group on all subdomain scores (t-tests: see Table I). A stepwise regression analysis inthe control group revealed a significant effect of sex(F (1, 191)=23.24, p<.001, males scoring higher),but no effect of age (t=)1.45, p=.149). Figure 1shows the Group and Sex differences graphically.
Table I. Mean AQ and Subscale Scores (and SDs) by Group
Communication Social Imagination Local details Attention switching Total AQ
Group 1
AS/HFA (n=52) x 8.2 7.8 6.7 6.1 8.5 37.3
SD 1.6 1.8 2.2 2.4 1.7 5.8
AS/HFA boys (n=38) x 7.9 7.4 6.8 6.1 8.2 36.4
SD 1.6 1.7 2.3 2.5 1.9 6.0
AS/HFA girls (n=14) x 9.0 8.8 6.8 6.1 9.0 39.8
SD 1.4 1.4 2.0 2.4 1.0 4.3
Group 2
Autism (n=79) x 8.0 8.0 7.6 6.5 8.3 38.3
SD 1.5 1.9 2.0 2.1 1.6 6.0
Autism boys (n=63) x 8.1 8.2 7.7 6.6 8.3 39.0
SD 1.5 1.8 2.0 2.2 1.6 5.9
Autism girls (n=16) x 7.6 7.3 6.8 5.9 8.1 35.7
SD 1.6 2.0 2.1 1.9 1.8 6.1
Group 3
Controls (n=50) x 2.7 2.0 3.2 5.3 4.5 17.7
SD 1.7 1.9 2.3 2.4 2.0 5.7
Control boys (n=25) x 2.9 2.2 4.4 5.8 5.0 20.2
SD 1.8 1.9 2.2 2.6 1.7 4.8
Control girls (n=25) x 2.6 1.8 2.0 4.8 4.1 15.3
SD 1.6 1.9 1.8 2.2 2.1 5.7
Controls vs. AS/HFA ta )16.9 )15.7 )7.9 )1.7# )10.7 )17.2Controls vs. autism ta )18.4 )17.7 )11.2 )2.8 )11.2 )19.2AS/HFA vs. autism Tb 0.7 )0.8 )2.0 )0.8 )0.7 )0.9
aAll t values in this row are significant at the p<.001 level except where shown (#).bNone of these t values are significant at the p<.05 level.
Adolescent AQ 345
An item analysis (percentage of each groupscoring on each item) is shown in Table II. On only 2items out of 50 (items 29, and 30) did controls scoremore than the clinical groups, strongly confirming thevalue of these items for discriminating autism spec-trum vs. controls. These two items were conserva-tively retained in the analysis since, if anything, theyserved to reduce the size of group differences. Theinternal consistency of items in each of the 5 domainswas also calculated, and Cronbach�s a Coefficientswere all high (Communication=0.82; Social=0.88;Imagination=0.81; Attention to Detail=0.66; Atten-tion Switching=0.76). Cronbach�s a Coefficient forthe AQ as a whole was also high (=0.79). Cronbach�sa Coefficients for each group were all in the range0.6–0.9.
The percentage of each group scoring at orabove each AQ score is shown in Table III. A usefulcut-off would discriminate the groups with as manytrue positives and as few false positives as possible. Inthe adults AQ, an AQ score of 32+ was chosen as auseful cut-off, since 79.3% of the AS/HFA groupscored at this level, whilst only 2% of controls did so.32+ also seemed to be a useful cut-off for distin-guishing adult females with AS/HFA (92.3% scoringat this point or above) vs. control adult females (1%of whom score at this point or above). Regarding theadolescent AQ, if the same cut-off score is used (of
32+), none of the control participants scored at thislevel, whilst all girls with AS/HFA scored above thislevel, as do 73.7% of the boys with AS/HFA. In theautism group, 68.8% of the girls, and 87.3% of theboys scored at this level. If we decrease the cut-offscore to 30+, none of the controls score above thislevel. All AS/HFA girls, and 86.8% of the AS/HFAboys score above this level, compared to 81.3% (girls)and 90.5% (boys) in the autism group. A cut-off atthis point might be considered in future screeningstudies.
Table III also shows that control females neverscore as high as 29+, whereas 4% of control malesdo. Note also that at AQ score 22+, there are almostfour times as many males (44%) as females (12%) inthe control group scoring at this intermediate pointon the scale. This suggests that there is not only a sexdifference on the child AQ overall (as reflected in themale mean AQ being higher than the female mean),and a sex difference at high levels on the AQ (reflectedin the sex ratio in Group 1 being 3.5:1), but thatsignificantly more males than females in the generalpopulation show moderate levels of ‘‘autistic traits’’.1
Again, this pattern of results replicates that foundwith the adult AQ. No differences were foundbetween the autism and HFA/AS groups on theAQ. Finally, test–retest reliability was high (r=.92,p<.001).
0
5
10
15
20
25
30
35
40
0 to
5
6 to
10
11 to
15
16 to
20
21 to
25
26 to
30
31 to
35
36 to
40
41 to
45
46 to
50
AQ score
Per
cen
t o
f su
bje
cts
Control girls
Control boys
AS/HFA/Aut boys and girls
Fig. 1. Percent of each group scoring at each group score.
346 Baron-Cohen, Hoekstra, Knickmeyer, and Wheelwright
DISCUSSION
In this paper, we report data from the adolescentversion of the Autism Spectrum Quotient (AQ), formeasuring the degree to which an individual adoles-cent shows autistic traits. As predicted, adolescentswith Asperger Syndrome (AS)/high functioning aut-ism (HFA) or with classic autism scored significantlyhigher on the AQ than matched controls. Eightypercent to 90% (mean=89.3%) scored above a criticalminimum of 30+, whereas none of the controls didso. This demonstrates that the adolescent AQ hasreasonable face validity, since the questionnaire pur-ports to measure autistic spectrum traits, and peoplewith a diagnosis involving these traits score highly onit. The adolescent AQ can also be said to havereasonable construct validity, in that items purportingto measure each of the 5 domains of interest (social,communication, imagination, attention to detail, andattention switching) show high a coefficients. Futurework needs to test the false negative rate. Theadolescent AQ has excellent test–retest reliability.
It is of interest that there were no significanteffects of age on adolescent AQ score, in the normalcontrol group. This suggests that what is beingmeasured by the AQ does not change with age, andthat the items are not biased towards one particularage group. Regarding the comparison of classicautism vs. HFA/AS, no significant differences werefound, which is also interesting. This may be becausethe items are not biased towards language skills.However, because of the communication subscale,which includes items about conversational compe-tence, we recommend that the AQ is primarily ofvalue for use with individuals with some speech, andwith an intelligence in the borderline average range(70) or above.
Within the control group, males score slightlybut significantly higher than females, both overall,and at intermediate and high levels of autistic traits.This is consistent with the extreme male brain theoryof autism (Asperger, 1944; Baron-Cohen, 2002;Baron-Cohen & Hammer, 1997) and may haveimplications for the marked sex ratio in autism andAS (Wing, 1981).
It is important to mention that this study wasnot in a position to compare the adolescent AQ toother instruments that have been developed tomeasure AS, such as the ASSQ (Ehlers, Gillberg, &Wing, 1999), the CAST (Scott, Baron-Cohen, Bol-ton, & Brayne, 2002), or the Australian Scale forAsperger Syndrome (Attwood, 1997). It will be of
Table II. Item Analysis: Percentage of Each Group Scoring on
Each Item
ItemAS/HFA (n=52)Autism (n=79)Controls (n=50)Subdomain
1 67.3 83.5 22.0 S
2 84.6 89.9 54.0 A
3 55.8 70.3 18.0 I
4 90.4 94.9 68.0 A
5 80.8 87.3 50.0 D
6 65.4 72.2 68.0 D
7 78.8 54.4 28.0 C
8 56.9 75.0 20.0 I
9 40.4 39.7 26.0 D
10 86.3 92.4 26.0 A
11 86.5 88.6 26.0 S
12 92.3 85.9 62.0 D
13 55.8 55.7 8.0 S
14 59.6 84.8 34.0 I
15 90.4 88.6 26.5 S
16 88.5 91.1 62.0 A
17 88.5 92.4 8.0 C
18 73.1 39.7 48.0 C
19 45.1 45.6 34.0 D
20 75.0 78.7 34.0 I
21 51.9 60.3 42.0 I
22 92.3 89.9 20.0 S
23 63.5 72.2 40.0 D
24 63.5 64.5 48.0 I
25 82.7 81.0 32.0 A
26 92.3 89.9 24.0 C
27 86.5 96.2 42.0 C
28 78.8 84.4 36.0 D
29 51.9 48.1 76.0 D
30 51.9 66.7 76.0 D
31 82.7 92.4 28.0 C
32 86.5 93.7 42.0 A
33 75.0 80.8 10.0 C
34 78.8 74.7 16.0 A
35 71.2 88.6 26.0 C
36 90.4 70.9 14.3 S
37 80.8 57.0 26.0 A
38 84.6 97.5 10.0 C
39 88.5 73.1 48.0 C
40 88.2 96.2 24.0 I
41 69.2 58.2 28.0 I
42 82.7 94.8 38.0 I
43 75.0 65.8 62.0 A
44 50.0 65.8 12.0 S
45 94.2 93.7 44.0 S
46 94.2 86.1 66.0 A
47 63.5 74.7 14.0 S
48 84.6 91.1 12.0 S
49 44.2 49.4 62.0 D
50 78.8 89.9 34.0 I
Key: S, Social skills; A, Attention switching; D, Attention to detail;
C, Communication; I, Imagination.
Adolescent AQ 347
interest in future studies to test how many cases ofAS each of these instruments correctly identifies (truepositive rate) as well as the rate of false negatives.Future work could also examine any relationship
between adolescent AQ score and severity ofsymptoms.
We wish to underline that the AQ is notdiagnostic, but may serve as a useful instrument in
Table III. Percent of Participants in Groups 1 and 2 Scoring at or Above each AQ Score—Overleaf
AQ
Score
AS/HFA
(n=52)
AS boys
(n=38)
AS/HFA
girls (n=14)
Autism
(n=79)
Autism boys
(n=63)
Autism girls
(n=16)
Controls
(n=50)
Control
boys (n=25)
Control
girls(n=25)
0 100 100 100 100 100 100 100 100 100
1 100 100 100 100 100 100 100 100 100
2 100 100 100 100 100 100 100 100 100
3 100 100 100 100 100 100 100 100 100
4 100 100 100 100 100 100 100 100 100
5 100 100 100 100 100 100 98.0 100 96.0
6 100 100 100 100 100 100 98.0 100 96.0
7 100 100 100 100 100 100 98.0 100 96.0
8 100 100 100 100 100 100 98.0 100 96.0
9 100 100 100 100 100 100 92.0 100 84.0
10 100 100 100 100 100 100 90.0 100 80.0
11 100 100 100 100 100 100 88.0 100 76.0
12 100 100 100 100 100 100 84.0 100 68.0
13 100 100 100 100 100 100 80.0 96.0 64.0
14 100 100 100 100 100 100 74.0 88.0 60.0
15 100 100 100 100 100 100 72.0 88.0 56.0
16 100 100 100 100 100 100 70.0 84.0 56.0
17 100 100 100 100 100 100 60.0 72.0 48.0
18 100 100 100 100 100 100 52.0 64.0 40.0
19 100 100 100 100 100 100 48.0 60.0 36.0
20 100 100 100 100 100 100 44.0 56.0 32.0
21 100 100 100 100 100 100 34.0 48.0 20.0
22 100 100 100 100 100 100 28.0 44.0 12.0
23 100 100 100 100 100 100 22.0 36.0 8.0
24 98.1 97.4 100 100 100 100 14.0 24.0 4.0
25 98.1 97.4 100 98.7 100 93.8 14.0 24.0 4.0
26 98.1 97.4 100 98.7 100 93.8 6.0 12.0 0
27 98.1 97.4 100 96.2 96.8 93.8 4.0 8.0 0
28 94.2 92.1 100 93.7 95.2 87.5 4.0 8.0 0
29 92.3 89.5 100 92.4 93.7 87.5 2.0 4.0 0
30 90.4 86.8 100 88.6 90.5 81.3 0 0 0
31 88.5 84.2 100 87.3 90.5 75.0 0 0 0
32 80.8 73.7 100 83.5 87.3 68.8 0 0 0
33 76.9 68.4 100 83.5 87.3 68.8 0 0 0
34 71.2 65.8 85.7 75.9 77.8 68.8 0 0 0
35 65.4 60.5 78.6 73.4 74.6 68.8 0 0 0
36 63.5 57.9 78.6 68.4 71.4 56.3 0 0 0
37 59.6 52.6 78.6 64.6 69.8 43.8 0 0 0
38 57.7 52.6 71.4 60.8 66.7 37.5 0 0 0
39 46.2 39.5 64.3 53.2 57.1 37.5 0 0 0
40 38.5 31.6 57.1 49.4 54.0 31.3 0 0 0
41 32.7 26.3 50.0 43.0 46.0 31.3 0 0 0
42 30.8 26.3 42.9 38.0 41.3 25.0 0 0 0
43 21.2 18.4 28.6 30.4 34.9 12.5 0 0 0
44 17.3 15.8 21.4 19.0 22.2 6.3 0 0 0
45 9.6 7.9 14.3 13.9 17.5 0 0 0 0
46 3.8 2.6 7.1 10.1 12.7 0 0 0 0
47 1.9 2.6 0 5.1 6.3 0 0 0 0
48 0 0 0 3.8 4.8 0 0 0 0
49 0 0 0 2.5 3.2 0 0 0 0
50 0 0 0 0 0 0 0 0 0
348 Baron-Cohen, Hoekstra, Knickmeyer, and Wheelwright
identifying the extent of autistic traits shown by aperson of normal intelligence. Currently this wouldbe for research purposes primarily, as the AQ hasnot been tested as a screening instrument in thegeneral population. Future research could comparethe sensitivity, specificity, and positive predictivevalue (PPV) of this instrument in communityscreening. Finally, the hope is that with quantitativeinstruments to measure the autistic spectrum acrossdifferent ages, this will improve comparability acrossresearch studies, assist in defining the phenotype ingenetic studies, and improve screening of undiag-nosed cases who need to be referred for a fulldiagnostic assessment.
NOTE
1. i.e. those traits which people with AS or HFAtend to endorse on the AQ.
ACKNOWLEDGMENTS
The authors were supported by a grant from theMRC and the Lurie-Marks Foundation, during theperiod of this work. Rosa Hoekstra was supported bya grant from the HsN Hersenstichting Nederland,the Bekker-La Bastide Fund and the SchuurmanSchimmel van Outeren Foundation.
APPENDIX 1: The Adolescent AQ
Definitely
agree
Slightly
agree
Slightly
disagree
Definitely
disagree
1. S/he prefers to do things with others rather than on her/his own.
2. S/he prefers to do things the same way over and over again.
3. If s/he tries to imagine something, s/he finds it very easy to create a picture in her/his mind.
4. S/he frequently gets so strongly absorbed in one thing that s/he loses sight of other things.
5. S/he often notices small sounds when others do not.
6. S/he usually notices car number plates or similar strings of information.
7. Other people frequently tell her/him that what s/he has said is impolite, even though s/he thinks
it is polite.
8. When s/he is reading a story, s/he can easily imagine what the characters might look like.
9. S/he is fascinated by dates.
10. In a social group, s/he can easily keep track of several different people’s conversations.
11. S/he finds social situations easy.
12. S/he tends to notice details that others do not.
13. S/he would rather go to a library than a party.
14. S/he finds making up stories easy.
15. S/he finds her/himself drawn more strongly to people than to things.
16. S/he tends to have very strong interests, which s/he gets upset about if s/he can’t pursue.
17. S/he enjoys social chit-chat.
18. When s/he talks, it isn’t always easy for others to get a word in edgeways.
19. S/he is fascinated by numbers.
20. When s/he is reading a story, s/he finds it difficult to work out the characters’ intentions.
21. S/he doesn’t particularly enjoy reading fiction.
22. S/he finds it hard to make new friends.
23. S/he notices patterns in things all the time.
24. S/he would rather go to the theatre than a museum.
25. It does not upset him/her if his/her daily routine is disturbed.
26. S/he frequently finds that s/he doesn’t know how to keep a conversation going.
27. S/he finds it easy to ‘‘read between the lines’’ when someone is talking to her/him.
28. S/he usually concentrates more on the whole picture, rather than the small details.
29. S/he is not very good at remembering phone numbers.
30. S/he doesn’t usually notice small changes in a situation, or a person’s appearance.
31. S/he knows how to tell if someone listening to him/her is getting bored.
32. S/he finds it easy to do more than one thing at once.
33. When s/he talks on the phone, s/he is not sure when it’s her/his turn to speak.
34. S/he enjoys doing things spontaneously.
35. S/he is often the last to understand the point of a joke.
36. S/he finds it easy to work out what someone is thinking or feeling just by looking at their face.
Adolescent AQ 349
REFERENCES
APA. (1994). DSM-IV diagnostic and statistical manual of mentaldisorders, 4th ed. Washington DC: American PsychiatricAssociation.
Asperger, H. (1944). Die ‘‘Autistischen Psychopathen’’ imKindesalter. Archiv fur Psychiatrie und Nervenkrankheiten,117, 76–136.
Attwood, T. (1997). Asperger�s Syndrome. UK: Jessica Kingsley.Baron-Cohen, S. (2002). The extreme male brain theory of autism.
Trends in Cognitive Sciences, 6, 248–254.Baron-Cohen, S., & Hammer, J. (1997). Is autism an extreme form
of the male brain?. Advances in Infancy Research, 11, 193–217.Baron-Cohen, S., Richler, J., Bisarya, D., Gurunathan, N., &
Wheelwright, S. (2003). The Systemising Quotient (SQ): Aninvestigation of adults with Asperger Syndrome or highfunctioning autism and normal sex differences. PhilosophicalTransactions of the Royal Society, Series B, Special issue on‘‘Autism: Mind and Brain’’, 358, 361–374.
Baron-Cohen, S., & Wheelwright, S. (2003). The FriendshipQuestionnaire (FQ): An investigation of adults with AspergerSyndrome or high functioning autism, and normal sex differ-ences. Journal of Autism and Developmental Disorders, 33, 509–517.
Baron-Cohen, S., & Wheelwright, S. (2004). The Empathy Quo-tient (EQ). An investigation of adults with Asperger Syndromeor high functioning autism, and normal sex differences. Jour-nal of Autism and Developmental Disorders, 34, 163–175.
Baron-Cohen, S., Wheelwright, S., Skinner, R., Martin, J., &Clubley, E. (2001). The Autism Spectrum Quotient (AQ):
Evidence from Asperger Syndrome/high functioning autism,males and females, scientists and mathematicians. Journal ofAutism and Developmental Disorders, 31, 5–17.
Ehlers, S., Gillberg, C., & Wing, L. (1999). A screening question-naire for Asperger Syndrome and other high-functioningautism spectrum disorders in school-age children. Journal ofAutism and Developmental Disorders, 29, 129–141.
Klin, A., Volkmar, F., Sparrow, S., Cicchetti, D., & Rourke, B.(1995). Validity and neuropsychological characterization ofAsperger Syndrome: Convergence with nonverbal learningdisabilities syndrome. Journal of Child Psychology and Psy-chiatry, 36, 1127–1140.
Scott, F., Baron-Cohen, S., Bolton, P., & Brayne, C. (2002). TheCAST (Childhood Asperger Syndrome Test): Preliminarydevelopment of UK screen for mainstream primary-schoolchildren. Autism, 6(1), 9–31.
Wakabayashi, A., Baron-Cohen, S., & Wheelwright, S. (2004). TheAutism-Spectrum Quotient (AQ) Japanese version: Evidencefrom high-functioning clinical group and normal adults. TheJapanese Journal of Psychology, 75, 78–84.
Wechsler, D. (1958). Sex differences in intelligence : The measure-ment and appraisal of adult intelligence. Baltimore: Williams &Wilking.
Wing, L. (1981). Asperger Syndrome: A clinical account. Psycho-logical Medicine, 11, 115–130.
Woodbury-Smith, M., Robinson, J., & Baron-Cohen, S. (2005).Screening adults for Asperger Syndrome using the AQ: diag-nostic validity in clinical practice. Journal of Autism andDevelopmental Disorders, 35, 331–335.
Appendix 1. (Continued)
Definitely
agree
Slightly
agree
Slightly
disagree
Definitely
disagree
37. If there is an interruption, s/he can switch back to what s/he was doing very quickly.
38. S/he is good at social chit-chat.
39. People often tell her/him that s/he keeps going on and on about the same thing.
40. When s/he was younger, s/he used to enjoy playing games involving pretending with other
children.
41. S/he likes to collect information about categories of things (e.g. types of car, types of bird, types
of train, types of plant, etc.).
42. S/he finds it difficult to imagine what it would be like to be someone else.
43. S/he likes to plan any activities s/he participates in carefully.
44. S/he enjoys social occasions.
45. S/he finds it difficult to work out people’s intentions.
46. New situations make him/her anxious.
47. S/he enjoys meeting new people.
48. S/he is a good diplomat.
49. S/he is not very good at remembering people’s date of birth.
50. S/he finds it very to easy to play games with children that involve pretending.
� MRC-SBC/SJW Feb 1998.
350 Baron-Cohen, Hoekstra, Knickmeyer, and Wheelwright